Language of African American Children: School Achievement and Vernacular Variants

1
Language of African American Children School
Achievement and Vernacular Variants
Joanne E. Roberts, PhD FPG Child Development
Institute The University of North Carolina at
Chapel Hill (UNC) Walt Wolfram, PhD North
Carolina State University Sandra C. Jackson, PhD
North Carolina Central University Jennifer
Renn, MA FPG Child Development Institute,
UNC Danai Kasambira, MAFPG Child Development
Institute, UNC 2006 American Speech-Language-Hear
ing Association Convention Miami Beach, FL
November, 2006

Research supported by National Science Foundation
(BCS-0544744) Maternal Child Health Bureau
(MCJ-370599, MCJ-379154 MCJ-370649,
R40MC-00343)
2
Acknowledgments

• Study children families
• Colleagues
• Funders National Science Foundation Maternal
  Child Health Bureau

3

• Please note pictures and data not published have
  been deleted from this version

4
Outline

• Issues and Measurement of AAE
• Comparison of Indices of AAE Style Shifting
• Study Rationale
• Design and Measures of AAE
• Study Results
• Child Family Risk Protective Factors for
  School Success
• Summary, Directions Discussion

5
Issues and Measurement of AAE
6
The Study of AAE

• Robust descriptive literature
• Selective descriptive representation
• The AAE cannon and empirical descriptive

7
The Supraregional Myth
Myth
Regionality in AAE is invariably trumped by the
trans-regional, common core of shared vernacular
traits.
8
The Supraregional Myth
Reality
Regionality played a significant role in the
earlier development of varieties of AAE and
continues to play a significant sociolinguistic
role in its construction.
9
The Language Trajectory Myth
Myth
There has been a unilateral trajectory of change
in AAE over the past century.
10
The Language Trajectory Myth
Reality
African American communities, particularly in the
rural South, may show alternative trajectories of
change with respect to core AAE structures and
regional accommodation, ranging from the
intensification of core AAE features concurrent
with the recession of regionalized features to
the dissipation of AAE features and the
maintenance of regionalized features.
11
The Social Stratification Myth
Myth
There is an isomorphic correlation between
demographically defined socio-economic status and
the use of acrolectal and basilectal AAE
12
The Social Stratification Myth
Reality
There are a host of community, contextual,
social, and personal factors that must be taken
into account in understanding the construction,
implementation, and regulation of vernacular and
mainstream norms in the African American
community.
13
Research Questions in AAE Development

• The developmental emergence of AAE
• Vernacularity in the life cycle
• Factors in the development and maintenance of AAE

14
Indexing and Measuring AAE

• Methods of measuring AAE use
• Dialect Density Measures (DDM)
• Selective predictability in measuring AAE use

15
Shifting Styles and AAE

• Maturation and dialect shift
• Situational sensitivity
• Selective structural shift
• Dialect control and manipulation

16
Dialect and Academic Achievement

• Dialect and school achievement
• Literacy and dialect
• AAE use in and out of school

17
Study Rationale Comparison of Indices of AAE
Style Shifting
18
Research Questions

• How effective are current methods (e.g., Dialect
  Density Measures) at quantifying style shift in
  AAE speakers?
• Can a subset of AAE features successfully account
  for vernacular use?
• How much sensitivity to situational context do
  adolescents demonstrate in their speech?
• What types of linguistic structures are utilized
  in effecting style shift?

19
Characterizing AAE

• Dozens of linguistic features can be considered
  characteristic of AAE
• Green (2002) specifies lexical, semantic,
  syntactic/morphosyntactic, and phonological
  features
• In most inventories morphosyntactic features
  greatly outnumber other categories
• Overlap with other varieties of English and
  regional difference can cause difficulties in
  characterizing AAE

20
AAE Feature Subsets

• Wolfram (1991) suggests that a subset of features
  might be diagnostic of AAE use
• A reduced feature list would be very useful
• Identifying several dozen features is extremely
  time consuming
• The full list of AAE features may not provide
  much more information than a carefully selected
  subset
• Reducing the number of features would allow for
  more advanced analysis methods
• The features selected might reveal important
  patterns about what is shifting

21
Study Method
22
Participant Characteristics Original Longitudinal
Study

• Data originally collected for longitudinal study
  of language development literacy in African
  American youth
• 88 longitudinal study participants recruited
• Entered study 6-12 months (M 8.1 months)

23
Participant Characteristics - Original
Longitudinal Study (Continued)

• Recruitment criteria
• Low to middle income African Americans
• Recruited from nine community childcare centers
• Children currently in 7th 10th grade

24
Overview of Study Design Longitudinal Sample

• Following children longitudinally from study
  entry (infancy) - 10th grade
• Repeated assessments - standardized tests,
  language samples, parent teacher reports
• Childrens language social skills (infancy
  10th grade) academic skills (4 yrs 10th
  grade) annually
• Peer relations (3rd 8th grade) annually
• Family environment (infancy 10th grade)
  annually until K, every 1 - 2 years K 10th
  grade
• Classroom/school (infancy 10th grade) annually

25
Grade 6 Peer Sample Procedures

• Each study participant recruited peer counterpart
  of same age gender in grade 6
• Peer sample similar to longitudinal sample in
  age, gender, grade in school, SES

26
Grade 6 Study Participants

• A subset of 50 study participants was randomly
  selected from overall sample
• Half longitudinal, half newly recruited
• 32 females, 18 males
• 12-14 years of age

27
Grade 6 Peer Procedures for Longitudinal Study

• Each dyad engaged in 2 formal 2 informal
  situations
• Formal contexts
• 2 mock speeches
• Informal contexts
• An issue discussion
• Free talk/snack period

28
Transcription Procedures

• Formal contexts - each speech was
  orthographically transcribed in its entirety
• Informal contexts - dialogue was transcribed
  until at least 50 communication units were
  attained for each subject

29
Coding Procedures

• 39 morphosyntactic 3 phonological features of
  AAE were coded
• Used Craig Washingtons (2006) Dialect Density
  Measure
• Also looked at 7 additional morphosyntactic
  features selected based on AAE literature
• Emphasis on morphosyntactic features
• More socially marked

30
Study Results Comparison of Indices of AAE
Style Shifting
31
Conclusions How Can We Measure Dialect Density?

• There was a high degree of reliability among all
  of the summary measures
• Using a selected subset of features seems to be
  diagnostic, especially when used specifically to
  quantify style shift
• Demonstrates close to the same degree of
  difference between contexts
• These 6 features may be especially useful because
  they are salient AAE features

32
Conclusions Does Style Shifting Occur?

• As expected, there was clear style shift between
  contexts with more AAE in INFORMAL regardless of
  the measure
• Speakers used a wider variety of vernacular
  features in informal situations
• Speakers used a lot of different features, but
  most were used rarely
• Suggests a knowledge of different features, but
  lack of use in style shifting

33
Child Family Risk Protective Factors for
School Success
34
Why Study School Success in African American
Youth?

• Many African American youth are very successful
  in school
• Important to identify characteristics of youth,
  family, school context that promote academic
  achievement school adjustment in African
  American youth

35
Risk and Achievement Gap In African American Youth

• Youth from families with multiple risk factors
  experience more school failure compared to
  children without risk factors
• African American youth as compared to white
  counterpart
• More likely to experience multiple risk factors
• Experience achievement gap at entry to
  kindergarten which widens during elementary
  middle school years

36
Study Purpose

• Does exposure to risk factors in early childhood
  relate to African American childrens school
  achievement adjustment?
• Do child family characteristics relate to
  better school outcomes despite exposure to social
  risk?

37
Factors Associated with Success Among African
American Children Exposed to Risk

• Protective factors - related to better outcomes
  in high-risk sample but not in a low-risk sample
  (interaction)
• Promotive factors - related to better outcomes
  for all children (main effects)
• Often thought to mediate anticipated negative
  pathway from exposure to multiple risk factors
  impaired academic trajectories
• High maternal education - promotive factor, low
  maternal education - risk factor

38
Longitudinal Study Population

• Recruited 88 African American children from 9
  community child care programs in 2 small southern
  cities
• Following children longitudinally from study
  entry (approximately 6 months) through 10th grade
• Repeated assessments - standardized tests,
  language samples, parent teacher reports
• Reporting on assessments through 6th grade
• 65 - 87 children in studies presenting today

39
Exposure to Risk Early Childhood (Burchinal et
al., 2006)
40
Proportion of Children with Early Risk Factors
at 1-4 Years
41
Social Risk Over Time

• Levels of social risk experienced in early
  childhood (6 months through K entry) highly
  correlated with social risk in G3 G6 (r .73 -
  .80)

42
Child Academic Outcomes Social Skills

• Outcomes K - G6 annually
• Reading math achievement scores
  Woodcock-Johnson Tests of Achievement (WJ)
• Social Skills Rating System
• Problem Behaviors

43
Early Childhood Family Risk Academic
Achievement in 2nd Grade
Low, Average, High Risk defined by one standard
deviation above or below the mean. Outcome is
W-Raush score (age equivalency measure)
44
Family Risk Problem Behaviors
45
Family Risk Factors as Predictors of School
Competence in K to G6

• Children with more family risk factors in early
  middle childhood showed less advanced skills in
  reading, math, social skills, more behavior
  problems from kindergarten entry through 6th grade

46
Mediators of School Successthrough Grade 6

• Do child language or parenting style serve as
  mediators or protective factors for African
  American children facing multiple risk factors?

47
Parenting Composite

• HOME (Caldwell Bradley, 1984) from infancy
  through 3rd grade
• Mothers teaching style
• Guessing Game mean communicative adequacy (clear
  specific cues) G1
• Magnet Task - explanatory talk ( science art
  talk) G1
• Principle component analysis

48
Child Language

• PPVT Receptive Vocabulary - K entry
• CELF-Receptive Language Expressive Language -
  K Entry Grades 2 4

49
Parenting as Protective Factor (Burchinal et al.,
(2006)
50
Language as Protective Factor for Math Skills
51
(No Transcript)
52
Conclusions Does Risk Relate to Achievement Do
Childrens Language Parenting Mediate School
Success?

• Children exposed to more risk factors in early
  childhood had lower school achievement
  adjustment
• Childrens language skills at K entry and in
  elementary school parenting served as
  protective factors in academic achievement of at
  risk children
• Language skills play integral role in academic
  achievement and adjustment, especially for
  children exposed to risk
• Other factors (e.g., childcare quality) during
  infancy/preschool served as protective factors
  for math problem behaviors in early elementary
  school

53
Summary, Directions, Discussion
54
The Broader Context of AAE

• Confronting the Diagnostic Issue
• Extending Developmental Description
• Documenting Dialect Shift Empirically

55
Shifting Styles in AAE

• Identifying shifting features
• Documenting and describing shift empirically
• Explaining the psycholinguistic and
  sociolinguistic basis of style shifting

56
AAE and Academic Achievement

• Identifying risk factors and AAE
• Language differences and academic development
• Language differences and speech and language
  assessment

57
Future Research AAE

• Complete sample 70 youth and 70 peers in grades 6
  8 in formal informal contexts
• Consider other ways to calculate DDMs other
  subsets to quantify AAE
• Determine whether different DDMs useful for
  diagnostic purposes
• Examine how other youth, family, peer, school
  factors affect AAE
• Examine relation of AAE school achievement when
  consider other youth, family, peer, school
  factors

58
Future Research Risk Protective Factors for
School Success

• High school follow-up 9th 10th grade - child,
  family school risk protective factors
• Studying peer relations, racial ethnic
  measures, working memory, executive function,
  narration
• Similar linkages in peer sample who are gender,
  grade, SES matches followed since 6th grade as
  in longitudinal sample

59
For More Information..

• Contact Joanne Robert--- joanne_roberts_at_unc.edu
• Research opportunities